22 CALIFORNIA ACADEMY OF SCIENCES 



Head broad and fiat, 3| to 3| in length; its depth at base of occipital process li to i§ in its 

 length. The dorsal profile from dorsal spine to tip of snout nearly (appearing perfectly) straight. 

 Snout wide and depresseti, sharply wedge-shaped as viewed from the side; its length to a point on the 

 median line of head between anterior margin of eyes 3| to 4 in head. The prefrontal does not form a 

 protuberance in front of eye as in G. jordani and related species, the contour of snout rising uniformly 

 to the edge of the dermal margin of eye. The interorbital area is flattened and wide; its width between 

 dermal supraorbital margins 2 to 2^ in head. The top of head is coarsely granular, the granulated area 

 extending forward usually to above middle of orbit in two diverging areas, separated by a triangular 

 continuation backward of the naked or smooth area of the snout. The length of the fontanel is 

 variable in the present specimens, but in none of them does it reach the occipital process.' In two 

 specimens evidently abnormal, the groove is very short, extending only to the ape.x of the triangular 

 smooth area, which extends backward from interorbital space. Usually, however, the groove narrows 

 backward uniformly, though sometimes constricted where it enters the granular area of the head, and 

 extends to within | diameter of eye of the occipital process. The occipital process is usually broader 

 than long; in extreme cases it is broader than long by nearly the length of the basal plate of the dorsal 

 spine; from this it varies to only as broad as long. The basal plate of the dorsal spine is twice, or a 

 little more than twice, as broad as long. The median keel is lower than in G. planiccps or G. jordani; 

 in some specimens it is scarcely evident at the termination of the fontanel, but grows stronger on the 

 occipital process. 



The maxillary barbel usually reaches base of pectoral spine; in one specimen to a.xillary pore, 

 in se\'eral about to middle of opercle. The outer mental barbels often reach but | distance to gill- 

 openings, but sometimes extend beyond the latter. The vomerine patches of teeth usually meet on 

 median line, or are separated by a narrow groove only. In but one specimen (an adult male) are 

 they widely separated (as figured by Giinther and Eigenmann). The eye is contained 6i to 7 times in 

 the head, measured in a straight line from median tip of snout obliquely above eye to gill-opening. 



As in other related species, the head is longer and the ventrals shorter in the male, the ventrals 

 overlapping front of anal in females and provided with a very large fold of the integument. In the 

 females the black of the ventrals passes gradually into the lighter margin. In males the transition 

 between the two areas is abrupt. In five specimens examined, the gill-rakers are 5-)-io, 5-|-io, 

 6+10, 5+ 1 1, and 5+12. The pectoral pore is a long slit. The base of adipose fin is twice or 

 nearly twice diameter of eye. The dorsal has 7 soft rays, and the anal 17 or 18. The length of 

 the pectoral is very variable; measuring from base of spine to tip of soft ra}'S, it is contained from 

 1 1 to i^ in head. 



Fi'om O. jordani jiiul other related sj)ecies, G. eigenmanni can be at once 

 recognized by the wider, flatter head, especially the more depressed snout, the smaller 

 eye, the rougher head, and the less evident carina on the occipital process. 



Three males and nine females were preserved. 



It is not clear what species from Santa Helena Bay is referred to by Boulenger 

 ( 1 <S!)8-91), Vol. Xlir, p. 5) under the name Arius seemnnni. From his statement 

 "Le GalelcJdhyn gUberti Jord., n'en est pas separable," we are disposed to believe that 

 he had before him specimens of G. jordani rather than G. eujenmanni. 



33. Galeichthys jordani {Eigenmann <fc Eigenmann). 



Two specimens were obtained 250 and 345 mm. long. The larger individual 

 is a female, with elongate ventrals overlapping the front of the anal fin, and provided 



' From a drawiug kimlly commiiuicated by Professor Garman, it is learned that Eigeumaim's material in 

 the Museum of Comparative Zoology agrees with the i^reseut specimens in the length of the fontanel groove. 



